U.S. patent number 5,027,641 [Application Number 07/520,084] was granted by the patent office on 1991-07-02 for oscillometric non-invasive blood pressure simulator.
Invention is credited to Leo F. Costello, Jr..
United States Patent |
5,027,641 |
Costello, Jr. |
July 2, 1991 |
Oscillometric non-invasive blood pressure simulator
Abstract
An apparatus for evaluating the performance characteristics of
an oscillometric blood pressure monitor is presented. In accordance
with the present invention, a novel closed-loop system is used to
synthesize physiologically-correct pneumatic pulse waveforms and
oscillometric envelopes in order to allow dynamic performance
evaluation of any oscillometric monitor, regardless of
manufacturer.
Inventors: |
Costello, Jr.; Leo F. (Hamden,
CT) |
Family
ID: |
26979431 |
Appl.
No.: |
07/520,084 |
Filed: |
May 7, 1990 |
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
314566 |
Feb 23, 1989 |
|
|
|
|
Current U.S.
Class: |
73/1.64 |
Current CPC
Class: |
A61B
5/0225 (20130101); G01L 27/005 (20130101); A61B
5/02225 (20130101) |
Current International
Class: |
A61B
5/0225 (20060101); G06F 17/00 (20060101); G01L
27/00 (20060101); G01L 027/00 () |
Field of
Search: |
;73/4R,865.6 ;128/681
;434/268 ;364/571.07,578 |
References Cited
[Referenced By]
U.S. Patent Documents
Other References
Biotek Product Literature-Blood Pressure Systems Calibrator Model
601A, Indirect Measurement in Man, pp. 86-99. Brochure-Critikon
Dinamap Vital Signs Monitor OG, Apr. 9, 1968, U.S. Pat. No.
3,376,660..
|
Primary Examiner: Raevis; Robert
Attorney, Agent or Firm: Fishman, Dionne & Cantor
Parent Case Text
This is a continuation of application Ser. No. 314,566, filed on
Feb. 23, 1989, now abandoned.
Claims
What is claimed is:
1. An apparatus for evaluating the performance characteristics of
an oscillometric blood pressure monitor which has a pressure sensor
and a pump and in which cuff pressure pulse signals are generated
when it is monitoring the blood pressure of a human subject,
comprising:
pulse waveform synthesizer means for synthesizing pulses with
predetermined waveforms that have shapes, slopes, amplitudes and
rates that are substantially the same as the corresponding
parameters of cuff pressure pulses taken from particular human
subjects and are therefore physiologically correct;
pulse generator means for converting said synthesized pulses to a
pressure signal of pulses which have substantially the same
parameters and physiological correctness as the synthesized pulses;
and
means for enabling said pressure signal to be sensed by the
pressure sensor in the monitor so that when the pump applies
pressure, the monitor will generate cuff pressure pulse signals
which have shapes, slopes, amplitudes and rates that are
substantially the same as would be generated by the monitor when
used to monitor the blood pressure of the human subjects whose cuff
pressure pulses are synthesized.
2. The apparatus of claim 1 including:
computer means which store blood pressure data taken from human
subjects, such data including the waveform parameters of shape,
slope, amplitude and pulse rate of the cuff pressure waveforms of
the subjects, said computer means being capable of delivering cuff
pressure waveforms which incorporate said parameters to the pulse
waveform synthesizer means.
3. The apparatus of claim 2 including:
control means for selecting from said computer means a
predetermined cuff pressure pulse waveforms.
4. The apparatus of claim 1 wherein the pump generates pressure
having a DC component and wherein said pulse generator means
generates pressure having an AC component, said apparatus further
including:
means for sensing the cuff pressure;
means communicating with said sensing means for separating the DC
and AC components of the sensed pump pressure; and
means responsive to the separating means for comparing the AC
component with the pulses from the pulse waveform synthesizer means
and delivering a correction signal to the pulse generator
means.
5. A simulator for testing an oscillometric blood pressure monitor,
said monitor being attachable to a living subject for determining
the blood pressure of the subject and having fluid pressure means
in which the pressure can be varied in the taking of blood pressure
measurements, and further having electrical means responsive to the
fluid pressure means for generating cuff pressure pulses of
characteristic shape from which the mean, systolic and diastolic
blood pressure of the subject are measured, comprising:
means for producing a series of fluid pressure pulses which have
waveforms with shapes, slopes and areas substantially the same as
the waveforms of the cuff pressure pulses of living subjects and
which incorporate the mean systolic and diastolic pressure and
pulse rate values of the subject, whereby the fluid pressure
waveforms are physiologically correct; and
means for transmitting the fluid pressure pulses to the fluid
pressure means of the monitor so that when the monitor is not
attached to a living subject and the pressure is varied by the
fluid pressure means, the electrical means will generate cuff
pressure pulses having substantially the same characteristic shape
as that generated when the monitor is checking the blood pressure
of a living subject, whereby the monitor can be tested by comparing
the values of blood pressure as measured by the monitor with the
values incorporated into the fluid pressure waveforms.
6. A simulator for testing an oscillometric blood pressure monitor
in which a fluid pressure source applies pressure to an inflatable
cuff attached to a living subject and a transducer senses the
pressure in the cuff and generates an electrical signal having cuff
pressure pulses of characteristic shape from which blood pressure
values of the living subject are determined, comprising:
electronic means which store in digitized form data on cuff
pressure waveforms collected from living subjects so that the
stored data defining a particular waveform are substantially the
same as the data defining the actual waveform collected from the
corresponding living subject and are thereby physiologically
correct, said data including the shapes, amplitudes and pulse rates
of the waveforms;
means for selecting stored waveforms by selecting a pulse rate and
amplitudes equal to desired systolic, mean and diastolic blood
pressures thereby to establish predetermined values for testing the
monitor;
means responsive to the selected waveforms for synthesizing analog
fluid pressure pulses having waveforms that incorporates the
predetermined values of the stored waveforms and is thus
physiologically correct; and
means for transmitting the fluid pressure pulses to the cuff so
that when said pressure source applies pressure to the cuff when
the cuff is not attached to a living subject, the transducer will
generate cuff pressure pulse signals of substantially the same
shape as that generated when the monitor is checking the blood
pressure of a living subject whereby the values of blood pressure
as measured by the monitor can be compared with the predetermined
values to test the monitor.
7. The simulator of claim 6 wherein the electronic means stores
data collected from human subjects having a variety of
cardiovasular conditions so that said selecting means can select
waveforms representative of particular cardiovascular conditions to
enable testing of the monitor ovar a range of types of human blood
pressure conditions.
8. A simulator for evaluating the accuracy of an oscillometric
blood pressure monitor attachable to a human subject for
determining blood pressure of the subject, said monitor having
fluid pressure means in which the pressure is increased and
decreased during the taking of the blood pressure of a human
subject, and having electrical means for generating cuff pressure
pulse signals of characteristic shape from which the mean, systolic
and diastolic blood pressures and the pulse rate of the human
subject are determined, comprising:
electronic means which store in digitized form data on cuff
pressure waveforms collected from human subjects so that the stored
data defining a particular waveform are substantially the same as
the data defining the actual waveform collected from the
corresponding human subject and are thereby physiologically
correct, said data including the shapes, amplitudes and pulse rates
of the waveforms;
means for selecting a stored waveforms by selecting a pulse rate
and amplitudes equal to desired systolic, mean and diastolic blood
pressures thereby to establish predetermined values for testing the
monitor;
means responsive to the selected waveforms for synthesizing analog
fluid pressure pulses having a waveforms that incorporates the
predetermined values of the stored waveforms and is thus
physiologically correct; and
means for transmitting the fluid pressure pulses to the fluid
pressure means of the monitor so that when the monitor is not
attached to a human subject and the pressure is increased and
decreased by the fluid pressure means, the electrical means will
generate cuff pressure pulse signals having substantially the same
characteristic shape as that generated when the monitor is checking
the blood pressure of a human subject, whereby the values of blood
pressure as measured by the monitor can be compared with the
predetermined values to evaluate the accuracy of the monitor.
9. An apparatus for evaluating the performance of an oscillometric
blood pressure monitor having a fluid pressure system including a
pressure source and a pressure sensor, comprising:
means for synthesizing predetermined waveforms;
pulse generator means for converting said predetermined waveforms
to a pressure signal;
means for transmitting said signal to the pressure system in an
oscillometric blood pressure monitor to produce a combined pressure
signal having AC and DC components when there is an applied
pressure in the system; and
means responsive to the combined signal for extracting its AC
component and for delivering to said pulse generator means a
correction signal based on a comparison of the waveforms of said AC
component and said predetermined waveforms.
10. A simulator for testing an oscillometric blood pressure monitor
which has fluid pressure means for varying the pressure during the
taking of blood pressure measurements and which also has electrical
means responsive to the fluid pressure means for generating cuff
pressure pulse signals from which the blood pressure of the subject
is measured, comprising:
means for producing a series of fluid pressure pulses which have
predetermined waveforms;
means for transmitting the fluid pressure pulses to the fluid
pressure means of the monitor so that when the fluid pressure means
varies the pressure a combined fluid pressure signal having AC and
DC components is generated;
means for separating the AC component from the combined signal;
and
means for developing a correction signal based on a comparison of
the waveform of the AC component and a preselected waveform and for
using the correction signal to modify the fluid pressure pulses at
the producing means.
11. A simulator for testing an oscillometric blood pressure monitor
which has fluid pressure means for varying the pressure during the
taking of blood pressure measurements and which also has electrical
means responsive to the fluid pressure means for generating cuff
pressure pulse signals from which the blood pressure of the subject
is measured, comprising:
means for producing a series of fluid pressure pulses which have
predetermined waveforms that have shapes, slopes, amplitudes and
rates which are substantially the same as the corresponding
parameters of the cuff pressure waveforms of living subjects from
whom blood pressure data has been collected;
means for transmitting the fluid pressure pulses to the fluid
pressure means of the monitor so that when the fluid pressure means
varies the pressure, a combined fluid pressure signal having AC and
DC components is generated;
means for separating the AC component from the combined signal,
said AC component having a first waveform; and
means for developing a correction signal based on a comparison of
said first waveform of the AC component and a preselected waveform,
said preselected waveform having first parameters associated
therewith, and means for using the correction signal to modify the
fluid pressure pulses at the producing means so that said
electrical means of the monitor will generate a substantially
distortion free cuff pressure pulse signal which incorporates said
first parameters of said preselected waveform.
12. An apparatus for evaluating the performance characteristics of
an oscillometric blood pressure monitor having a pressure sensor
and a pump comprising:
pulse waveform synthesizer means for synthesizing predetermined
waveforms to define a predetermined oscillometric envelope
pattern;
pulse generator means for converting said predetermined pulse
waveforms to a pressure signal capable of being sensed by the
pressure sensor in the oscillometric blood pressure monitor;
pressure transducer means for sensing pressure produced by the pump
in the oscillometric blood pressure monitor;
high pass filter means electrically communicating with said
pressure transducer means, said high pass filter means adapted to
separate the DC and AC components from the sensed pump pressure;
and
error amplifier means for receiving the AC component from said high
pass filter means, said AC component having a first waveform, said
error amplifier means comparing a preselected waveform with said
first waveform of said AC component sensed by said pressure
transducer means and delivering a correction signal to said pulse
generator means.
13. In a method for testing an oscillometric blood pressure monitor
which produces a characteristic oscillometric envelope signal from
which systolic, mean and diastolic blood pressures and pulse rate
of a living subject are determined when the monitor is connected to
the subject, said method involving the transmission of simulated
cuff pressure pulses to the monitor, including the steps of:
collecting blood pressure and cuff pressure data from living
subjects which includes blood pressure waveforms with their unique
parameters of shapes, slopes, areas, rise and fall times, rates and
amplitudes, and obtaining systeolic, mean, diastolic pressures and
pulse rate via direct measurement, so that the data define
physiologically correct waveforms of the subjects;
electronically storing the collected blood pressure data;
selecting a desired waveform for testing such a monitor by
selecting certain systolic, mean and diastolic pressures and a
pulse rate; and
converting the selected waveform into fluid pressure pulses having
substantially the same waveform and incorporating substantially the
same parameters as the electronically stored waveform, for
transmission to such a monitor so that the monitor will produce its
characteristic oscillometric envelope signal incorporating said
unique parameters in substantially the same manner as if the
monitor were being used to monitor the blood pressure of the living
subject from whom the stored data was collected.
14. In a method for testing an oscillometric blood pressure monitor
which produces a characteristic oscillometric envelope signal from
which systolic, mean and diastolic blood pressures and pulse rate
of a living subject are determined when the monitor is connected to
the subject, said method involving the tansmission of simulated
cuff pressure pulses to the monitor, including the steps of:
collecting blood pressure and cuff pressure data from living
subjects which includes cuff pressure waveforms with their unique
parameters of amplitudes, shapes, slopes, areas, rise and fall
time, and rates, so that the data define physiologically correct
waveforms of the subjects;
electronically storing the collected blood pressure data;
selecting a desired waveform for testing such a monitor by
selecting certain systolic, mean and diastolic pressures and a
pulse rate;
converting the selected waveform into fluid pressure pulses having
substantially the same waveform and incorporating substantially the
same parameters as the electronically stored waveform for
transmission to such a monitor to generate a combined pressure
signal having AC and DC components;
developing a correction signal based on said AC component; and
applying the correction signal so as to counteract distortion in
the fluid pressure pulses.
15. A method for evaluating the performance characteristics of an
oscillometric blood pressure monitor having a pressure sensor and a
pump comprising the steps of:
synthesizing predetermined waveforms to define a predetermined
oscillometric envelope pattern;
converting said predetermined pulse waveforms to a pressure signal
capable of being sensed by the pressure sensor in the oscillometric
blood pressure monitor;
sensing pressure in the oscillometric blood pressure monitor, said
pressure being sensed by pressure transducer means;
using high pass filter means to separate the DC and AC components
from the sensed pump pressure; and
using error amplifier means for receiving the AC component from
said high pass filter means, said AC component having a first
waveform, said error amplifier means comparing a preselected
waveform with said first waveform of said AC component sensed by
said pressure transducer means and delivering a correction signal
to said pulse generator means.
Description
BACKGROUND OF THE INVENTION
This invention relates generally to the field of automated
measurement of blood pressure by the oscillometric method. More
particularly, this invention relates to a method and apparatus for
evaluating and testing the performance characteristics of
oscillometric blood pressure monitors and associated cuff.
Oscillometry has become the preferred method for non-invasive blood
pressure (NIBP) monitoring. Oscillometric NIBP monitors
(oscillometers) are widely used in the Operating Room, and are
increasingly used as an alternative to invasive monitoring in the
Intensive Care Unit and Coronary Care Unit.
Commercially available oscillometric monitors have proprietary
algorithms for predicting the mean, systolic and diastolic pressure
from the oscillometric envelope. The more advanced monitors also
use complex algorithms intended to reduce noise such as motion
artifact. However, no systematic, standardized method is available
to evaluate the efficiency of these new devices and algorithms.
Static calibration does not test dynamic functionality, human
subject comparisons are unreliable and clinical studies are not
suitable for routine calibration and performance verification.
U.S. Pat. No. 4,464,123 to Glover et al describes a device for
evaluating oscillometers. However, the oscillometer evaluator of
the Glover et al patent is capable only of evaluating a specific
line of oscillometers manufactured by a specific company under a
very specific and artificial set of conditions. The Glover et al
device may not yield meaningful test data for oscillometers
manufactured by other companies.
U.S. Pat. No. 4,189,936 to Ellis discloses a device for testing
invasive blood pressure monitors as opposed to non-invasive
oscillometers.
SUMMARY OF THE INVENTION
The above-discussed deficiencies and other drawbacks of the prior
art are overcome or alleviated by the method and apparatus of the
present invention for evaluating the performance characteristics of
an oscillometric blood pressure monitor. In accordance with the
present invention, a novel closed-loop system is utilized to
synthesize physiologically-correct pneumatic pulse waveforms and
oscillatory envelopes in order to allow dynamic performance
evaluation of any oscillometric monitor, regardless of
manufacturer.
The present invention comprises a non-invasive blood pressure
simulator which includes a cylindrical housing for receiving the
cuff from the oscillometer to be tested. The housing encloses a
pulse waveform synthesizer which generates a desired continuous
electrical analog pulse waveform which is fed to a pneumatic or
hydraulic pulse generator. A pressure transducer then senses the
cuff pressure.
An important feature of the present invention is a servo loop for
maintaining correct signals. This servo loop comprises a high pass
filter and an error amplifier wherein the cuff pressure at a
pressure transducer in the oscillometer is sensed by the pressure
transducer in the simulator and its DC or steady state component is
stripped away by the high pass filter; and its AC component is fed
back to the error amplifier. The error amplifier compares the
desired pulse waveform with the actual waveform and drives the
pulse generator with a correction signal.
The above discussed and other features and advantages of the
present invention will be appreciated and understood by those of
ordinary skill in the art from the following detailed description
and drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
Referring now to the drawings, wherein like elements are numbered
alike in the several FIGURES:
FIG. 1 is a graphical depiction of blood pressure (BP) versus
time;
FIG. 2 is a graphical depiction of cuff pressure versus time as
measured by an oscillometric blood pressure monitor;
FIG. 3 is a graphical depiction of cuff pulse amplitude versus mean
cuff pressure as measured by an oscillometric blood pressure
monitor;
FIG. 4 is a perspective view of a non-invasive blood pressure
simulator in accordance with the present invention shown connected
to an oscillometer to be measured;
FIG. 5A is a schematic of the assembly of FIG. 4;
FIG. 5B is a schematic of an alternative embodiment of the present
invention;
FIG. 5C is a schematic of still another alternative embodiment of
the present invention; and
FIG. 6 is a block diagram of a typical clinical data collection
set-up for obtaining physiologically correct BP waveforms.
DESCRIPTION OF THE PREFERRED EMBODIMENT
Prior to describing the structural details of the present
invention, a brief discussion of non-invasive blood pressure
monitoring, particularly monitoring using oscillometry techniques
now follows. With reference first to FIG. 1, the arterial blood
pressure waveform has four parameters that are of clinical
interest:
1) The diastolic pressure is the minimum pressure during a single
beat (cardiac cycle).
2) The mean arterial pressure (MAP) is the average pressure during
a single cardiac cycle.
3) The systolic pressure is the maximum pressure during a single
cardiac cycle.
4) The heart rate, in beats per minute, can be computed from the
time interval, or period, from one cardiac cycle to the next.
Note that, in general, all four parameters will vary on a
beat-to-beat basis.
These four arterial blood pressure parameters may be measured
indirectly by the auscultatory method (the familiar procedure that
uses a stethoscope and arm cuff), the ultrasound method (an
ultrasonic blood flow detector replaces the stethoscope) and the
oscillometric method. The first two methods, auscultatory and
ultrasound, measure only the diastolic and systolic pressures. The
latter method, oscillometry, actually measures only the mean
arterial pressure, and estimates the systolic and diastolic
pressures, as will be discussed below. All three methods measure
the heart rate.
Oscillometry employs an occluding cuff identical to that used by
nurses when manually taking a subject's blood pressure by the
auscultatary method. However, unlike the auscultatary procedure,
oscillometry determines the blood pressure entirely from the cuff
pressure without the need for other patent sensors such as a
stethoscope, microphone or ultrasound probe.
FIG. 2 illustrates a time plot of the cuff pressure obtained during
a typical oscillometric measurement cycle. The cuff is inflated
above the subject's systolic pressure and then step-wise decreased
below the diastolic pressure. When the cuff pressure is decreased
to some pressure close to but still greater than the systolic
pressure, each heartbeat will superimpose a pulse (or oscillation)
on the cuff pressure. As the cuff pressure is decreased to the mean
pressure, the pulses gradually increase to a maximum. As the cuff
pressure is further decreased past the diastolic, the pulses
gradually grow smaller and eventually disappear altogether.
FIG. 3 shows these pulse amplitudes plotted against the mean cuff
pressure at which they occurred. This plot is known as the
oscillometric envelope. Clinical studies and theoretical
investigations have established that the maximum pulse amplitude
usually occurs at the mean blood pressure. Thus, to measure the
MAP, oscillometric blood pressure instruments record this
oscillometric curve, determine its maximum and read the
corresponding cuff pressure.
In contrast, systolic and diastolic pressures are derived from the
oscillometric envelope in a much less exact manner. In all
oscillometric monitors commercially available at present, this is
done by looking up the pressures above and below the mean at fixed
relations of the maximum amplitude. No precise relationship between
the systolic and diastolic pressures and the mean has been
established. As a result, new oscillometric monitor designs are
empirically "calibrated" to determine the best relationship. In
fact, procedures which are based upon the derivative of the
oscillometric envelope instead of the fixed fractional relations
have been proposed as offering greater accuracy.
Aside from the systolic and diastolic pressures accuracy problem,
the other practical problem with oscillometry is the reliable
detection of the cuff pressure pulses. Noise can be introduced into
the cuff pressure from, for example, arm motion or vibrations (in
an ambulance or helicopter). Oscillometer manufacturers employ a
variety of proprietary pulse detection algorithms to reject or
cancel such noise.
The present invention comprises a simulator which interacts with
the BP monitor it is testing (the "monitor-under-test") so as to
generate oscillometric pulse waveforms and an oscillometric
pulse-amplitude envelope that are standardized, repeatable and
physiologically correct. An important feature of the simulator of
the present invention is that it is suitable for use with
oscillometers from any manufacturer. The MAP, systolic BP,
diastolic BP and heart rate are "encoded" in the oscillometric
envelope pattern. Thus, the user can select these four parameters
and the present invention will synthesize cuff pulses whose
amplitudes define an envelope corresponding to the selected blood
pressures and whose spacing yields the selected heart rate. The
measurements obtained by the monitor-under-test can then be
compared to the known true values.
The shape of the synthesized pulses and envelope, the pulse periods
and the values for the MAP, systolic and diastolic pressures are
all based upon a database obtained from clinical studies of the
populations at large (see discussion of FIGURE 6). Thus, the
monitor-under-test will receive physiologically-correct
signals.
FIG. 4 depicts a preferred embodiment of the NIBP simulator of this
invention. The simulator 10 is housed in a cylindrical enclosure 12
with a control console 14 (containing switches and an alphanumeric
readout) on the top face 16 of enclosure 12. The cylindrical shape
of closure 12 provides a convenient post for wrapping an arm cuff
18 from an oscillometer 20 to be tested. A pair of tubes 22 and 23
extending from the base of cylindrical enclosure 12 allows
simulator 10 to interconnect with the oscillometric monitor 20 to
be tested. A "T" fitting 24 at the monitor end of tubes 22 and 23
is plugged into the tubing receptacle at the rear of monitor 20.
Tubes 22 and 23 are then plugged into fitting 24. A pair of second
and third tubes 26 and 27 are connected between fitting 24 and cuff
18. "T" fitting 24 allows monitor 20 to communicate with arm cuff
18 and also allows simulator 10 access to the cuff pressure.
FIG. 5A is a block diagram of a preferred embodiment of simulator
10 with an attached monitor-under-test 20. Simulator 10 comprises a
pulse waveform synthesizer 28, an error amplifier 30, a pneumatic
pulse generator 32, a high pass filter 34 and a pneumatic pressure
transducer 36. Simulator 10 also includes the pair of tubes 22 and
23 and fitting 24. A micro computing unit 41 delivers control
information to the pulse waveform synthesizer 28 from control
display unit 14. (Note however, that micro computing unit 41 may be
an integral part of pulse waveform synthesizer 28). The known
monitor-under-test components include a pressure transducer or
sensor 38, a fluid pump 40, the oscillometer cuff 18 and the cuff
tubing 26 and 27. Note that a dual lumen tube constitute the cuff
tubing 26 and 27 with a flow lumen for connecting the pump to the
cuff, and a sense lumen for connecting the cuff to the pressure
sensor. Tubing 22 interconnects fitting 24 with pneumatic pressure
transducer 36 and acts to deliver the cuff pressure to the
simulator 10. Tubing 23 interconnects fitting 24 with pneumatic
pulse generator 32 and acts to deliver the pulse waveform to
oscillometer 20. In the interest of clarity, a substantial portion
of the monitor-under-test, such as pneumatic valves and control and
signal processing functions are not shown. However, such monitors
are well known and are disclosed in detail in, for example, U.S.
Pat. No. 4,349,034.
The components function as follows: The microcomputing unit 41
delivers to pulse waveform synthesizer (PWS) 28 numerical samples
of the desired pulse waveform. PWS 28 then converts these numbers
into a continuous electrical analog of the desired. waveform.
Assuming for the moment that the feedback from the high pass filter
(HPF) 34 is zero, pneumatic pulse generator (PPG) 32 is driven by,
in effect, the output of PWS 28. PPG 32 then converts this
electrical analog into the desired pneumatic signal.
Because of non-linearities in PPG 32, or unpredictable loading by
the oscillometer cuff and tubing, the actual pneumatic pulse
waveform arriving at oscillometer pressure sensor 38 may deviate
from the desired waveform. In anticipation of this, the cuff
pressure at pressure sensor 38 is sensed by pneumatic pressure
transducer 36 and its "DC" or steady-state component is stripped
away by HPF 34 and its "AC" component is fed back to error
amplifier 30. Error amplifier 30 compares the desired pulse
waveform with the actual and drives PPG 32 with a correction signal
thus defining a servo loop.
The "servo loop" just described is an important feature of the
present invention since electromechanical transducers (such as PPG
32) are well known for non-linearities and difficulties in
maintaining precision over time.
Two factors in particular will likely contribute to distortion of
the desired pulse shape:
1) The mean, or DC, cuff pressure may vary from 30 mmHg to almost
300 mmHg. PPG 32 is directly exposed to this pressure; so it must
superimpose the pulse waveform on top of it.
2) PPG 32 is operating under a load that is beyond control of the
simulator system. Since simulator 10 is a universal device suitable
for testing oscillometric monitors from many different
manufacturers, variation in the properties (compliance and
stiffness) of the oscillometer cuff and tubing must be allowed
for.
Aside from PPG 32, the electronic components depicted in simulator
10 of FIG. 5A are all well known and commercially available. The
feedback network of pneumatic pressure transducer (PPT) 36 and high
pass filter 34 should be stable and modifiable. In a preferred
embodiment, PPT 36 comprises a piezoelectric integrated circuit
pressure sensor such as Model SX05 manufactured by Sensym of
Sunnyvale, Calif. This PPT responds to DC pressures of up to 350
mmHg and has a frequency response that begins to roll off at about
10,000 Hertz. It will be appreciated that such a piezoelectric
silicon diaphragm will yield a very stable transducer. HPF 34
preferably comprises a fourth order Bessell filter. A preferred
error amplifier 30 comprises a Linear Technologies, LT1013.
Simulator control console 14 will comprise any known suitable
membrane keyboard and suitable display means such as LCD or LED
types. Microcomputer 41 and PWS 28 preferably comprise a
commercially available Motorola 68HCll in conjunction with a
suitable digital-to-analog (D/A converter). It will be appreciated
that both HPF 34 and error amplifier 30 may comprise software as
opposed to electronic hardware.
The purpose of the Pneumatic Pulse Generator (PPG) (and the
Hydraulic Pulse Generator which will be discussed hereinafter) is
to transform the electrical pulse waveform into an equivalent
fluidic pressure signal. Thus, the PPG is a type of transducer
since it converts energy from one form (electrical) to another
(mechanical).
While the PPG may consist of many different embodiments, in the
preferred embodiment, this energy conversion is accomplished as a
two step process: First, an electro-mechanical linear motion
transducer converts the electrical signal into a corresponding
mechanical displacement. Second, this displacement is converted to
a fluidic pressure by the piston in a pneumatic or hydraulic
cylinder.
While the linear motion transducer could consist of a variety of
devices such as a conventional "snap-acting" linear solenoid or a
digital linear actuator (a stepper motor fitted with a lead screw),
the preferred embodiment is a "proportional" linear solenoid such
as a Lucas Ledex Model PS-16. The PS-16 provides a mechanical
displacement, or stroke, that is proportional to input current. To
convert this displacement to a pressure change, the present
invention couples the plunger, or shaft, of the PS-16 to the shaft
of a suitable pneumatic/hydraulic cylinder such as a Clippard Model
7SD-1 miniature cylinder. The cylinder functions like a piston pump
in reverse: Displacement of the cylinder shaft moves the cylinder
piston against the fluid (either air or liquid) thereby compressing
the fluid and converting the displacement into a fluid pressure
change.
Operation of the simulator of the present invention begins by
selecting the desired MAP, systolic and diastolic BP and heart rate
on the control console 14. Pneumatic cuff 18 of the
monitor-under-test 20 is then wrapped around simulator body 12 and
the measurement cycle begins. Assuming the monitor has an integral
pump, it inflates the cuff to the initial pressure and then begins
deflating the cuff and looking for pressure pulses. Simulator 10
senses the cuff pressure via tubing 22 and 27; and derives
corresponding pulse amplitudes from the selected envelope pattern.
Pressure pulses having these amplitudes are generated at the
desired heart rate and are transmitted to the cuff 18 via tubing
23. Thus, as monitor 20 deflates cuff 18, the pulses it measures
will be those carefully synthesized by simulator 10 in order to
create the desired envelope pattern. A two-way "dialogue" between
simulator 10 and monitor 20 takes place: monitor 20 controls the DC
cuff pressure at any given time and simulator 10 senses it and
responds with oscillometric pulses whose amplitude is a function of
the DC cuff pressure.
FIGS. 5B and 5C depict less preferred embodiments of the NIBP
simulator of the present invention with structure corresponding to
that found in FIG. 5A having the same reference numeral. The
simulator 10' of FIG. 5B differs from FIG. 5A in that the pulse
waveforms are hydraulically generated rather than pneumatically
generated. In FIG. 5B, pulse waveform synthesizer 28 delivers
signals to a hydraulic pulse generator 50 which communicates with a
hydraulic bladder 52. Hydraulic bladder 52 thus acts to deliver
simulated pressure pulses to oscillometer cuff 18. Hydraulic
pressure transducer 54 senses the DC component or mean pressure
from bladder 52 while the cuff 18 senses the AC component or pulse
wave form from bladder 52.
FIG. 5B retains the important "servo loop" feature of FIG. 5A. That
is, the embodiment of FIG. 5B is a closed-looped system which
incorporates feedback to sense and control the pulse waveform at
the output of tubing 22. In the FIG. 5B embodiment, the connection
between simulator 10' is tapped at tubing 22. The tapped pressure
is sensed by the pneumatic pressure transducer 36, its DC component
is stripped by the high-pass filter 34 and the resulting AC
pressure signal is fed back to the error amplifier 30. The error
amplifier 30 then compares the actual pressure signal with the
desired signal from pulse wave form synthesizer 28 and drives the
hydraulic pulse generator 50 with a correction signal.
The simulator 10" of FIG. 5C depicts still another embodiment o the
present invention which is less preferred than either the
embodiments of FIGS. 5A or 5B. Basically, the FIG. 5C embodiment is
an open-looped simulator, that is, FIG. 5C is analogous to the FIG.
5B embodiment without the "servo-loop" feature. This FIG. 5C
embodiment of simulator 10" includes a hydraulic pulse generator
50, hydraulic pressure transducer 54 and hydraulic bladder 52. As
in FIG. 5B, bladder 52 is wrapped about cylindrical housing 12 and
the oscillometer cuff 18 is wrapped about bladder 52. During a
simulated measurement cycle, the mean or DC pressure in the cuff is
transferred to the bladder 52 where it can be sensed by transducer
54. Meanwhile, the pulse or AC pressure waveforms synthesized by
generator 50 are transmitted via bladder 52 to cuff 18.
FIG. 5C includes the significant feature of the present invention
wherein the simulator 10" is completely "non-invasive" to the
oscillometer-under-test. This allows for testing of the
oscillometer without the need for dismantling or otherwise
disturbing the oscillometer. Unfortunately, while this embodiment
is indeed non-invasive; it does not contain the important
"servo-loop" feature which adversely impacts on the accuracy of the
simulator as discussed above.
Another important feature of the present invention is the
physiological correctness of the synthesized pulse waveforms and
oscillometric envelope. Because the pulse waveforms are
physiologically correct, simulator 10 can provide a valid "test
suite" to evaluate the effectiveness of pulse detection algorithms.
In fact, measured amounts of noise (of various types) could be
superimposed on the pulses to allow algorithm behavior to be
studied in a controlled manner.
While the physiologically correct synthesized pulse waveforms and
oscillometric envelope may be determined by any desired method, a
preferred technique for obtaining such data is through a clinical
study such as follows.
The primary concern in this empirical (clinical) study is to
collect waveforms that are representative of what an oscillometric
BP monitor will sense. The clinical data collection set-up for
characterizing the cuff pulse waveforms is presented in FIG. 6.
Direct arterial pressures are recorded from the left (preferably)
radial artery and, when possible, from the aorta. The invasive
measurements are obtained via catheter-tip pressure transducers
(Medical Measurements, Hakensack, N.J.) in order to eliminate the
artifact often caused by "catheter whip". The transducer signals
are fed to the patient monitor (Model VR-16, Honeywell E for M,
Pleasantville, N.Y.) which will isolate, amplify and display the
pressures. The patient monitor also feeds the two pressure signals
to the Analog I/O Board (Model DT2808, Data Translation, Marlboro,
Mass.) where they are digitized to 12 bit resolution at 50
samples/second and stored in the hard disk of the microcomputer
(Portable II, Compaq, Richardson, Tex.). The microcomputer also
displays the acquired digitized signals in order to provide a
quality check of the recordings.
A Critikon Dynamap cuff (Model 1845, Critikon, Tampa, Fla.) sized
to the patient according to AHA guidelines (cuff width
approximately 40% of arm circumference) is applied to the right arm
(contralateral to the catherized arm). The standard Dynamap 8 foot,
dual-lumen cuff tube is used to connect the cuff to the pressure
transducer within the oscillometer interface (custom built).
The output of this transducer is split into DC and AC components by
filters in the oscillometer interface. It is essential that the
bandwidth of the AC component is as wide as is necessary in order
not to distort the pulse waveform.
The DC and AC cuff pressure signals are then digitized to 12 bit
resolution by the analog I/0 board. The AC pressure component is
sampled at 100 samples/second (assuming this is consistent with the
bandwidth needed) while the DC component is sampled at 10
samples/second; both digitized signals are stored in the hard
disk.
The flow-lumen of the cuff tubing connects the cuff to the
inflation pump and deflation valves with the oscillometer
interface. The microcomputer sends signals to the oscillometer
interface to control the cuff inflation, deflation and venting. The
data collection and cuff control software is written using the
ASYST scientific data acquisition and analysis software
package.
The following protocol is used to record the four signals:
1) Enter patient clinical history into a database via the
microcomputer.
2) Measure the brachial pressure in each arm via manual
auscultation.
3) Catheterize the patient and verify the stability and quality of
the direct pressure signal(s) on the patient monitor display and on
the PC.
4) Apply the arm cuff to the subject and ask him/her to remain
quiet.
5) Initiate a "test inflation" cycle via the PC and the
oscillometer interface and verify the stability and quality of the
two indirect pressure signals on the PC. (This will inflate the
cuff to about 100 mmHg for about 10 sec.).
Initiate a "data collection" cycle and record the four signals for
60 seconds. A data collection cycle consists of a rapid inflation
to well above systolic pressure, cuff deflation until pulses occur,
deflation at 5 mmHg steps and pausing for 2 cardiac cycles and
rapid venting after pulses disappear. A 60 second cycle avoids
problems related to venous pooling.
This procedure is repeated four times for each the subjects. Each
recording session generates a data file, 25 KiloBytes long. These
files are stored on the data collection microcomputer's hard disk
and transferred periodically via floppy disk and/or modem, to the
signal processing workstation at another location. The data files
(approximately 8 megabytes in total) are archived on the
workstation hard drive and used for the data analysis phase of the
project.
Next, the data is analyzed. The object of the data analysis is to
identify the typical characteristics of the pulse waveform and
oscillometric envelope in subjects with and without peripheral
vascular disease. The characteristics that are analyzed are listed
below:
Pulse waveform characteristics:
width and duty cycle
rise time
fall time
energy (area under the curve)
maximum and minimum slopes
waveshape indices
The last item needs clarification: An indication of the general
similarity of waveform shapes at different mean cuff pressures is
obtained by computing the autocorrelation function (ACF) of the
pulse waveform for a given subject during a single data collection
cycle. To access waveform shape similarity at the same mean cuff
pressures for a given subject, all pulse waveforms collected for a
given mean cuff pressure will be "waveform summed". The average of
the summed waveforms are used as the template or "kernel" to
compute a crosscorrelation function (XCF) that will indicate
waveshape similarities. Finally, the waveshapes for pulses at the
same cuff pressure in different subjects are compared by computing
the XCF of each subject's template waveform and the template of
every other subject.
Four envelope characteristics include:
(1) location of arterial diastolic, mean and systolic pressures
(2) characteristic ratios for systolic, diastolic and mean
(3) characteristic slopes for systolic, diastolic and mean
(4) waveshape indices
The waveshape indices for the envelope will be obtained similarly
to those for the pulse waveform except that, obviously, the ACF
cannot be computed.
The signal analysis is performed by software written in ASYST,
which provides interactive signal plotting, numerical analysis and
statistics. The software analyzes each set of four acquired signals
as follows:
1) Each of the four signals is displayed and the signal quality is
checked visually.
2) For the arterial pressure signal(s), the following parameters
are computed on a beat-to-beat basis and their mean and standard
deviations calculated:
mean pressure
systolic pressure
diastolic pressure
pulse pressure (systolic-diastolic)
systolic fraction ("duty cycle")
waveform coefficient ([mean-dia]/[sys-dia])
3) The pulse waveform parameters listed above (except for the
waveshape indices) are computed for all the pulses collected at the
various mean pressure values.
4) The envelope parameters (except for the waveshape indices) are
computed from the mean values computed in step 2).
This procedure is repeated for each of the four sets collected for
every patient. If the pulse and envelope parameters derived from
these sets appear to be consistent they are averaged to form the
patient parameter set. If not, the patient data will be
disregarded. This entire process is repeated for the data from each
patient and then the waveshape indices are computed.
Multi-linear regression and standard statistical analysis
techniques are used to correlate the dependent, oscillometric
variables--the pulse and envelope parameters--with the independent,
physiological variables--the arterial BP parameters and the
cardiovascular condition of the subject.
Physiologically-correct pulse waveforms and oscillometric envelopes
(derived from the clinical testing described in detail above)
allows for valid testing of any oscillometric monitor regardless of
the particular method used to determine systolic and diastolic
pressures which is not possible with prior art devices such as that
disclosed in previously mentioned U.S. Pat. No. 4,464,123 to Glover
et al. The Glover et al device uses an oscillometric envelope which
is a mathematically convenient, three segment linearized function;
and can only be used with the oscillometer of a particular
manufacturer for obtaining meaningful test results. For example,
another manufacturer may have selected different fractional
relationships (which provide the best accuracy for that specific
oscillometer implementation); or, an entirely new approach (such as
the aforementioned derivative method) not based on fractional
relations may be employed by a given manufacturer. In contrast, the
present invention, because it recreates the true physiological
waveforms, will provide valid results with either of these latter
two cases, whereas the device of Glover et al would be useless.
The NIBP simulator in accordance with the present invention allows
complete dynamic evaluation of any monitor (regardless of
manufacturer) under conditions that emulate an actual measurement
by using a novel preferably closed loop system to synthesize
physiologically correct, pneumatic pulse waveforms and oscillatory
envelopes in order to allow dynamic performance evaluation of any
oscillometric monitor.
The simulator of this invention will improve the accuracy of
oscillometric BP monitors currently used and it will encourage
refinements in oscillometric technology that will lead to more
accurate BP monitor designs. Specifically, it will provide:
1) a standardized signal source for comparing oscillometers from
any manufacturer;
2) a means for independent validation of oscillometer performance
claims;
3) a quick, objective method for clinical staff to verify
oscillometer performance without placing a service call to
biomedical engineering;
4) a test fixture that will simulate motion artifact and other
types of noise in order to evaluate oscillometer performance under
adverse conditions;
5) a method to analyze the accuracy of the different systolic and
diastolic blood pressure prediction algorithms used in
oscillometers; and
6) a data base of clinically derived oscillometric pulse waveforms
and envelopes representing various cardiovascular conditions that
can be used to develop improved algorithms for pulse detection,
noise reduction and systolic/diastolic prediction.
As is clear from FIG. 5A, the simulator of the present invention
may be used without oscillometer cuff 18. In this case, a reservoir
would be provided in simulator 10 to act as a container for
retaining fluid which would have been received in cuff 18.
Alternatively, simulator 10 could include suitable venting means
which would vent that fluid normally delivered to the cuff. Thus,
the simulator of the present invention could, in effect, bypass the
cuff 18 and concentrate solely on the accuracy and operation of the
oscillometer under test.
While an important feature of the present invention is to allow for
evaluation of the instrument (electronic) section of an NIBP
monitor, another application of this invention is to allow
evaluation of the oscillometer cuff and cuff tubing. Cuff and
tubing properties of interest include:
1. the transfer function (i.e., the input/output response);
2. the impedance (pressure change per flow rate change);
3. stiffness (pressure change per volume change);
4. compliance (flow rate change per pressure change per second);
and
5. hermetic integrity (airtightness).
It will be apPreciated that the impedance, stiffness and compliance
are the pneumatic analogs of, respectively, the electrical
quantities resistance, capacitance and inductance.
While preferred embodiments have been shown and described, various
modifications and substitutions may be made thereto without
departing from the spirit and scope of the invention. Accordingly,
it is to be understood that the present invention has been
described by way of illustrations and not limitation.
* * * * *